Data warehouse development for public sentiment from social media towards tourism in Yogyakarta

As users increasingly turn to YouTube and TikTok for travel inspiration and information, these platforms have emerged as key hubs where travelers share their journeys and seek recommendations from fellow users. Amid this surge in user-generated content, a significant challenge arises from the vast amount of unstructured data embedded in comments and reactions. This unstructured data lacks a predefined structure, making it difficult to analyze using traditional methods. However, it holds immense potential for gaining insights into traveler preferences, sentiment, and emerging trends. This study addresses this challenge by using DIY Yogyakarta as data source, the most favored holiday destination in 2022. The primary objective is to aggregate and analyze public sentiment towards tourism in this region using data from YouTube and TikTok. Employing advanced sentiment analysis techniques, specifically the Naïve Bayes Classifier, which is well-suited for text classification due to its simplicity, efficiency, and high accuracy in handling large datasets, the study achieves an accuracy of 90% in sentiment analysis. The ETL (Extract, Transform, Load) process is meticulously designed to integrate data from TikTok and YouTube, complemented by Google Maps Tourism data, into a MySQL database, ensuring data integrity and consistency. The resulting dashboard provides actionable insights into public sentiment dynamics, platform effectiveness, sentiment trends over time, and influential user contributions. These insights are poised to inform strategic content strategies, optimize influencer engagement, and monitor emerging tourism trends, thereby enhancing tourism services and planning initiatives.